a) Metabolic MRI to profile prostate cancer to target glycolytic pathways: Hyperpolarized (HP) MRI using pyruvate allows imaging kinetics of important bio-energetic pathways such as aerobic glycolysis or oxidative phosphorylation. Since malignant cells process glucose/pyruvate through aerobic glycolysis vs normal cells which rely on oxidative phosphorylation, this imaging allowed biochemically profiling regions of interest in vivo non-invasively using 13C MRI using hyperpolarized tracers. Hyperpolarization bridges the 4-orders in sensitivity needed for 13C MRI in vivo. Two human prostate cancer cell lines (DU145 and PC3) were grown as xenografts. The conversion of pyruvate to lactate in xenografts was measured with hyperpolarized [1-13C]-pyruvate MRSI after systemic delivery of [1-13C] pyruvic acid. Steady state metabolomic analysis of xenograft tumors was performed with mass spectrometry and steady state lactate concentrations were measured with proton (1H) MRSI. Tumor growth was assessed after lactate dehydrogenase (LDH) inhibition with FX-11. DU145 tumors demonstrated an enhanced conversion of pyruvate to lactate with hyperpolarized [1-13C]-pyruvate MRSI compared to PC3, and a corresponding greater sensitivity to LDH inhibition. We showed that hyperpolarized [1-13C]-pyruvate MRSI magnetic resonance spectroscopic imaging (MRSI) of prostate cancer predicts efficacy of targeting the Warburg effect. b) Synthesis and evaluation of 13C-labeled 5-5-dimethyl-1-pyrroline-N-oxide aimed at in vivo detection of reactive oxygen species using hyperpolarized 13C-MRI: 5,5-Dimethyl-1-pyrroline-N-oxide (DMPO) is a spin trap agent frequently used to detect reactive oxygen species (ROS) in vitro. In this study, we synthesized 13C-labeled DMPO to be applied to hyperpolarized 13C-MRI in which 13C MR signal increases more than 10000 folds, and investigated feasibility of in vivo ROS detection by the 13C-labeled DMPO combined with hyperpolarized 13C-MRI. DMPO was 13C-labeled at C5 position, and deuterated to prolong the T1 relaxation time. Overall yield achieved for 5-13C-DMPO-d9 was 15%. Hyperpolarized 5-13C-DMPO-d9 provided a single peak at 76 ppm on the 13C-spectrum, and the T1 relaxation time was 60 sec in phosphate buffer. The solution of hyperpolarized 5-13C-DMPO-d9 was injected into a mouse placed in a 3T scanner, and 13C-spectra was acquired every 1 sec. The signal of 5-13C-DMPO-d9 was detected in the living mouse body, and the T1 decay of 13C signal of hyperpolarized 5-13C-DMPO-d9 in the mouse body was 29 sec. 13C-chemical shift imaging revealed that 5-13C-DMPO-d9 was distributed through the mouse body in a minute after the intravenous injection. The strong signal of 5-13C-DMPO-d9 was detected in heart/lung and kidney, whereas the signal in liver was small compared to other organs. The results indicate hyperpolarized 5-13C-DMPO-d9 provided sufficient magnitude of the 13C signal to be detected in the mouse body, and can be applied to some disease models to evaluate the capability for detection of ROS in vivo. c) Molecular imaging of the microenvironment of pancreatic tumor xenografts in mice guides treatment strategy with radiotherapy- or hypoxia-activated prodrugs: Pancreatic ductal adenocarcinoma (PDAC) is characterized by hypoxic niches, leading to treatment resistance. Therefore, studies of tumor oxygenation and metabolic profiling will contribute to improved treatment strategies. Here, we demonstrated the ability of two imaging biomarkers to predict differences in tumor response to therapy: 1) partial oxygen pressure (pO2), measured by EPR imaging; and 2) [1-13C] pyruvate metabolism rate, measured by hyperpolarized 13C MRI. Three human PDAC xenografts with varying treatment sensitivity (Hs766t, MiaPaCa-2, and Su.86.86) were grown in mice. The median pO2 of the mature Hs766t, MiaPaCa-2, and Su.86.86 tumors was 9.1, 11.1, and 17.6mmHg, and the rate of pyruvate-to-lactate conversion was 2.72, 2.28, and 1.98 min-1, respectively(n=6, each). The results are in agreement with steady state data of matabolites quantitatively measured by mass spectroscopy and histological analysis indicating glycolytic and hypoxic profile in Hs766t and MiaPaca-2 tumors. Fractionated radiation therapy (5 Gy x 5) resulted in a tumor growth delay of 16. and 18. days in MiaPaca-2 and Su.86.86 tumors, respectively, compared to 6. days in hypoxic Hs766t tumors. Treatment with gemcitabine, a first-line chemotherapy, or the hypoxia-activated prodrug TH-302 was more effective against Hs766t tumors (20. and 25. days' increase in survival time, respectively) than MiaPaCa-2 (2. and 6. days) and Su.86.86 (4. and 0.7 days) tumors. Collectively, these results demonstrate the ability of molecular imaging biomarkers to predict the response of PDAC to treatment with radiation therapy and TH-302.